WO2023244184A1 - Quality inspection system for rice using artificial intelligence technology - Google Patents

Quality inspection system for rice using artificial intelligence technology Download PDF

Info

Publication number
WO2023244184A1
WO2023244184A1 PCT/TH2023/050008 TH2023050008W WO2023244184A1 WO 2023244184 A1 WO2023244184 A1 WO 2023244184A1 TH 2023050008 W TH2023050008 W TH 2023050008W WO 2023244184 A1 WO2023244184 A1 WO 2023244184A1
Authority
WO
WIPO (PCT)
Prior art keywords
rice
unit
images
rice grain
instance segmentation
Prior art date
Application number
PCT/TH2023/050008
Other languages
French (fr)
Inventor
Phuvin KONGSAWAT
Nawapat JAMROENRAK
Wisuwat SUNHEM
Nuttakan WIRIYAKRIENG
Matichon MANEEGARD
Sujitra SUTTHITHATIP
Original Assignee
Easyrice Digital Technology Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Easyrice Digital Technology Co., Ltd. filed Critical Easyrice Digital Technology Co., Ltd.
Publication of WO2023244184A1 publication Critical patent/WO2023244184A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06395—Quality analysis or management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30128—Food products

Definitions

  • Rice quality inspection involves various factors to consider. For example, it includes determining whether each grain is broken or intact, as well as sorting rice into different categories. This process begins with randomly selecting rice samples and then categorising them into types such as Whole kernels, Brokens, New crop, Chalky kernels, or Paddy. These categories are then calculated as proportions or percentages based on weight.
  • this inspection process has limitations, including the time-consuming nature of evaluating each sample and the requirement for trained inspectors.
  • different personnel may have varying standards for inspection. Therefore, researchers and innovators have conducted studies and developments to address the issues of delays and inaccuracies in rice quality inspection. The following are examples of such innovations that have emerged.
  • Thai Patent Publication No. 124484 entitled “Method for calculating weight ratio by quality in the external quality classifier of grains”, presents a method for calculating weight ratios by differentiating the quality of grain seeds through pixel-based calculations. The method involves multiplying the pixel count of a captured image by the weight coefficient per pixel, which is obtained from a large dataset.
  • the invention does not mention the use of artificial intelligence technology or learning techniques.
  • Thai Patent Publication No. 131022 entitled “Digital image separation method of oval grains”, demonstrates the separation and classification of adjacent seeds of cereal crops by manipulating the pixels on the outer edges in all direction until only the central axis pixels, representing the arrangement of the seeds, remain. These remaining pixels are then compared to the framework lines to facilitate seed separation.
  • the invention does not employ colour-based seed differentiation techniques, and the technique may not accurately separate or distinguish overlapping seeds.
  • the quality inspection system for rice using artificial intelligence technology comprises a data transmission and display unit, which includes a scanner responsible for capturing images of rice grains and sending them to a computer.
  • the aforementioned computer then sends the scanned images to a server, which creates a network and handles data exchange from a cloudbased processing unit.
  • the server allocates resources and scheduling the access to the instance segmentation and attribute extraction of rice grains unit. It also segments and identifies positions of rice grains and extracts the attributes of the grains.
  • the cloud-based processing unit also responsible for segmenting and identifying the positions of rice grains and extracts the attributes of the rice grains using a gradient-based learning system to instance segmenting and identifying the position of each rice grain.
  • This learning system utilises scanned images of rice grains from the scanner to learn and generate images of segmented and attributed rice samples. It accomplishes this by learning from human-labelled example images and subsequently applying this knowledge to segment and identify physical attributes.
  • the objective of this invention is to develop a quality inspection system for rice using artificial intelligence technology, capable of accurately and swiftly segmenting individual rice grains even when they are placed closely together.
  • This system enables precise and rapid assessment of rice quality, thereby overcoming the limitations of the rice industry in terms of quality inspection, reducing the time required for inspection, and minimising the risk of errors.
  • the main purpose is to use this system for inspecting rice quality according to the standards of Thai rice and sample-based rice purchasing, which requires random sampling from paddy and white rice in both the purchasing and selling processes.
  • Figure 1 Display one of the designs of a scanner that is suitable for use in the quality inspection system for rice using artificial intelligence technology.
  • Figure 2 Demonstrates a particular representation of the overall configuration of the quality inspection system for rice using artificial intelligence technology.
  • Figure 3 Exhibits one aspect of the learning method employed by the quality inspection system for rice using artificial intelligence technology.
  • Figure 4 Showcase one aspect of the training method and the gradient-based learning system.
  • the quality inspection system for rice using artificial intelligence technology demonstrates one particular configuration of the rice quality inspection system using artificial intelligence technology. Its purpose is to assess rice quality according to the standards of Thai rice and sample-based rice purchasing, which requires random sampling from paddy and white rice in both the purchasing and selling processes.
  • Figures 1 to 4 depict the characteristics of the quality inspection system for rice using artificial intelligence technology. This system consists of the following components:
  • the data transmission and display unit are equipped with a flatbed scanner, such as the one shown in Figure 1, which has a black background to produce suitable images.
  • This example scanner has a width of 250 millimeters, a length of 367 millimeters, and a thickness of 42 millimeters, capable of capturing images at a maximum resolution of 2400 dpi. Its function is to receive images data and send it to the computer and the internet connected computer, which includes at least a display monitor.
  • the computer system connected to the internet receives sample rice grain images from the scanner and forwards them to the server for further processing;
  • the server consisting of a resource allocation and scheduling unit, is responsible for receiving data from the computer to execute resource allocation and data scheduling to the instance segmentation and attribute extraction of rice grains unit. This unit is used to segment and identify the position of rice grains and extract their attributes.
  • the resource allocation and scheduling unit determines which data are to access to the instance segmentation and attribute extraction of rice grain unit before or after, and whether to access the instance segmentation and attribute extraction of rice grain unit located on the server or the cloud-based processing unit; -
  • the cloud-based processing unit ensures the presence of, at least, an instance segmentation and attribute extraction of rice gain unit.
  • the cloud-based processing unit serves as an images database for the end-to-end gradient-based learning system to learn. It accumulates a large amount of learning data, resulting in making the outcome from end-to-end gradient-based learning system more accurate compared to other learning methods.
  • the resource allocation and scheduling unit can be located on the server or the cloudbased processing unit, or both. However, the most suitable configuration is to have it present on both the server and the cloud-based processing unit.
  • the end-to-end gradient-based learning system learns from either human-labelled example images, or images obtained from database located on the cloud-based processing unit. It then uses these images to identify physical attributes.
  • the system employs the following components: a. The instance segmentation separating background pixels from rice grain pixels unit. b. The instance segmentation separating adjacent rice grain pixels unit. c. The inference accessing attributes of rice grain unit, at least.
  • the process begins with scanning sample rice grains. Once the rice grains are scanned using a scanner device, the resulting images are sent to a computer connected to the internet. Subsequently, the computer sends the received image data to the server for resource allocation and scheduling to access the instance segmentation and attribute extraction of rice grain unit, which may be located either on the server or in the cloud-based processing unit.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Marketing (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
  • Image Analysis (AREA)

Abstract

This invention application demonstrates a quality inspection system for rice using artificial intelligence technology. It comprises a data transformation and display unit that sends sample images for processing at the cloud-based processing unit. The server receives data from a computer to execute resource allocation and data scheduling to process in the instance segmentation and attribute extraction of rice grain unit. This unit is responsible for segmenting and identifying the positions of rice grains, as well as extracting their attributes. The cloud-based processing unit is also involved in processing and includes the end-to-end gradient-based learning system that consists of the instance segmentation separating background pixels from rice grain pixels unit, the instance segmentation separating adjacent rice grain pixels unit, and the inference assessing the attributes of rice grains unit. The system aims to obtain images of segmented and attributed rice samples.

Description

QUALITY INSPECTION SYSTEM FOR RICE USING ARTIFICIAL
INTELLIGENCE TECHNOLOGY
Field of the Invention
Engineering, specifically in the part related to the quality inspection system for rice using artificial intelligence technology.
Background of the Invention
In Thailand and abroad, the quality inspection of rice is a highly significant process due to its role in assessing the standards of rice products. This process is crucial for rice mills, rice exporters, or rice ports in determining pricing and evaluating the purchase or export of rice. For example, when exporting Thai Hom Mali rice to the other countries, the rice must undergo quality inspection according to the standards set by a ministry or relevant government agencies. Therefore, the quality inspection of rice must be accurate, precise, clear, and fast, considering the large quantity of rice products. Additionally, quality inspection of rice is also crucial in the rice production process, as it requires the improvement of rice quality or production methods to meet the standards set by the ministry. Currently, the rice industry in Thailand faces the following main problems in the rice purchasing process:
Issues regarding delays in rice quality inspection arise from the method of measurement that relies on skilled personnel with knowledge of rice quality. The process involves randomly selecting a 25-gram sample per ton of produced rice, which is then used for quality assurance. In this sample, there are approximately 1,200 rice grains. However, in the current human analysis-based inspection format, the size and attributes of each grain can only be assessed once, requiring 1-4 hours to evaluate the quality of the rice product. The duration depends on the expertise of the personnel. This problem has led to a bottleneck in the Thai rice industry.
In addition to the issue of delays, there are also problems arising from human errors due to the fact that skilled personnel have to work for 8 hours a day. This led to fatigue, resulting in inaccuracies in evaluating the quality of rice product that do not align with the actual quality of the produced rice. Consequently, this leads to rejected shipments and tarnishes the reputation of Thai rice, causing widespread repercussions.
Small-scale community rice mills and small rice mills do not have the necessary tools to assets the quality of rice products in accordance with the standards set by the Ministry of Commerce. Quality assurance of rice requires personnel with expertise, but community rice mills and small rice mills lack sufficient budget to hire such personnel. As a result, the rice products they produce lack quality assessment, leading to low-grade products and lower prices in the market.
Rice quality inspection involves various factors to consider. For example, it includes determining whether each grain is broken or intact, as well as sorting rice into different categories. This process begins with randomly selecting rice samples and then categorising them into types such as Whole kernels, Brokens, New crop, Chalky kernels, or Paddy. These categories are then calculated as proportions or percentages based on weight. However, this inspection process has limitations, including the time-consuming nature of evaluating each sample and the requirement for trained inspectors. Furthermore, different personnel may have varying standards for inspection. Therefore, researchers and innovators have conducted studies and developments to address the issues of delays and inaccuracies in rice quality inspection. The following are examples of such innovations that have emerged.
Thai Patent No. 77414, entitles “Methods and systems for non-degradation of rice varieties”, is an invention that proposes a method and system for classifying rice varieties based on paddy or white rice using optical knowledge. However, the invention described in the aforementioned Thai patent does not involve data transmission through computers or the use of artificial intelligence technology.
Thai Patent Publication No. 116799, entitled “Machine and method for classification of grain quality using photos”, is an invention related to quality inspection of flat-shaped grain seeds through the use of photographs that capture the characteristics of the grain seeds from all sides. This invention proposes a methodology for inspecting the quality of grain seeds from all sides. The invention proposes a methodology for inspecting the quality of grain seeds but does not specifically mention the use of artificial intelligence technology.
Thai Patent Publication No. 144633, entitled “Optical characterisation device”, is an invention that presents a method and measurement system for differentiating the properties of objects with seed-like shapes, such as plant seeds or pills, using the principles of light and images analysis. However, it does not discuss the integration of artificial intelligence technology in this operation.
Thai Patent Publication No. 124484, entitled “Method for calculating weight ratio by quality in the external quality classifier of grains”, presents a method for calculating weight ratios by differentiating the quality of grain seeds through pixel-based calculations. The method involves multiplying the pixel count of a captured image by the weight coefficient per pixel, which is obtained from a large dataset. However, the invention does not mention the use of artificial intelligence technology or learning techniques. Thai Patent Publication No. 131022, entitled “Digital image separation method of oval grains”, demonstrates the separation and classification of adjacent seeds of cereal crops by manipulating the pixels on the outer edges in all direction until only the central axis pixels, representing the arrangement of the seeds, remain. These remaining pixels are then compared to the framework lines to facilitate seed separation. The invention does not employ colour-based seed differentiation techniques, and the technique may not accurately separate or distinguish overlapping seeds.
Summary of the Invention
The quality inspection system for rice using artificial intelligence technology comprises a data transmission and display unit, which includes a scanner responsible for capturing images of rice grains and sending them to a computer. The aforementioned computer then sends the scanned images to a server, which creates a network and handles data exchange from a cloudbased processing unit. Additionally, the server allocates resources and scheduling the access to the instance segmentation and attribute extraction of rice grains unit. It also segments and identifies positions of rice grains and extracts the attributes of the grains. The cloud-based processing unit also responsible for segmenting and identifying the positions of rice grains and extracts the attributes of the rice grains using a gradient-based learning system to instance segmenting and identifying the position of each rice grain. This learning system utilises scanned images of rice grains from the scanner to learn and generate images of segmented and attributed rice samples. It accomplishes this by learning from human-labelled example images and subsequently applying this knowledge to segment and identify physical attributes.
The objective of this invention is to develop a quality inspection system for rice using artificial intelligence technology, capable of accurately and swiftly segmenting individual rice grains even when they are placed closely together. This system enables precise and rapid assessment of rice quality, thereby overcoming the limitations of the rice industry in terms of quality inspection, reducing the time required for inspection, and minimising the risk of errors. The main purpose is to use this system for inspecting rice quality according to the standards of Thai rice and sample-based rice purchasing, which requires random sampling from paddy and white rice in both the purchasing and selling processes. Brief Description of the Drawings
Figure 1: Display one of the designs of a scanner that is suitable for use in the quality inspection system for rice using artificial intelligence technology.
Figure 2: Demonstrates a particular representation of the overall configuration of the quality inspection system for rice using artificial intelligence technology.
Figure 3: Exhibits one aspect of the learning method employed by the quality inspection system for rice using artificial intelligence technology.
Figure 4: Showcase one aspect of the training method and the gradient-based learning system.
Detailed Description of the Invention
The quality inspection system for rice using artificial intelligence technology, as presented as this invention, demonstrates one particular configuration of the rice quality inspection system using artificial intelligence technology. Its purpose is to assess rice quality according to the standards of Thai rice and sample-based rice purchasing, which requires random sampling from paddy and white rice in both the purchasing and selling processes.
Figures 1 to 4 depict the characteristics of the quality inspection system for rice using artificial intelligence technology. This system consists of the following components:
- The data transmission and display unit are equipped with a flatbed scanner, such as the one shown in Figure 1, which has a black background to produce suitable images. This example scanner has a width of 250 millimeters, a length of 367 millimeters, and a thickness of 42 millimeters, capable of capturing images at a maximum resolution of 2400 dpi. Its function is to receive images data and send it to the computer and the internet connected computer, which includes at least a display monitor. The computer system connected to the internet receives sample rice grain images from the scanner and forwards them to the server for further processing;
- The server, consisting of a resource allocation and scheduling unit, is responsible for receiving data from the computer to execute resource allocation and data scheduling to the instance segmentation and attribute extraction of rice grains unit. This unit is used to segment and identify the position of rice grains and extract their attributes. The resource allocation and scheduling unit determines which data are to access to the instance segmentation and attribute extraction of rice grain unit before or after, and whether to access the instance segmentation and attribute extraction of rice grain unit located on the server or the cloud-based processing unit; - The cloud-based processing unit ensures the presence of, at least, an instance segmentation and attribute extraction of rice gain unit. It performs segmenting and identifying the position of each rice grain, as well as extracting their attributes using an end-to-end gradientbased learning system for segmenting and identifying position of each rice grain based on the images captured by the scanner, resulting in images of segmented and attributed rice samples. Additionally, the cloud-based processing unit serves as an images database for the end-to-end gradient-based learning system to learn. It accumulates a large amount of learning data, resulting in making the outcome from end-to-end gradient-based learning system more accurate compared to other learning methods.
The resource allocation and scheduling unit can be located on the server or the cloudbased processing unit, or both. However, the most suitable configuration is to have it present on both the server and the cloud-based processing unit.
The end-to-end gradient-based learning system learns from either human-labelled example images, or images obtained from database located on the cloud-based processing unit. It then uses these images to identify physical attributes.
To achieve physical attributes identification, the system employs the following components: a. The instance segmentation separating background pixels from rice grain pixels unit. b. The instance segmentation separating adjacent rice grain pixels unit. c. The inference accessing attributes of rice grain unit, at least.
To obtain images of segmented and attributed rice samples by using the quality inspection system for rice using artificial intelligence technology, the process begins with scanning sample rice grains. Once the rice grains are scanned using a scanner device, the resulting images are sent to a computer connected to the internet. Subsequently, the computer sends the received image data to the server for resource allocation and scheduling to access the instance segmentation and attribute extraction of rice grain unit, which may be located either on the server or in the cloud-based processing unit.
Best Mode of the Invention
As described in Detailed Description of the Invention

Claims

Claims Quality inspection system for rice using artificial intelligence technology, consisting of: Data Transmission and display unit (1);
Server (2);
Cloud-based processing unit (3); characterised in that: the data transmission and display unit (1) consists of a scanner (12) that scans rice grains images and sends them to a computer (11), which in turn sends the scanned images to a server; the server (2) consists of a resources allocation and scheduling unit (21) and an instance segmentation and attribute extraction of rice grain unit (22) which receive data from the computer to execute resource allocation and data scheduling to process in the instance segmentation and attribute extraction of rice grain unit to segment and extract attributes of the rice grains; the cloud-based processing unit (3) includes an instance segmentation and attribute extraction of rice grain unit (21) and a database (31) as a minimum requirement which store sample rice grain images, segment and identifies the position of the rice grains, and extract their attributes using the end-to-end gradient-based learning system (311) to segment and identify position of each rice grain using sample rice grain images (310) for learning and generates images of segmented and attributed rice samples (6) and sends them back to the data transmission and display unit for display; the end-to-end gradient-based learning system learns from human-labelled example images (4), and then takes these human-labelled example images and identifies physical attribute (51); identify physical attributes (51) by organising them into: a. the instance segmentation separating background pixels from rice grain pixels unit (510); b. the instance segmentation separating adjacent rice grain pixels unit (511); and c. the inference assessing the attributes of rice grains unit (512); these units are the minimum requirements essential for generating images of segmented and attributed rice samples (6). The quality inspection system for rice using artificial intelligence technology, according to the principles stated in claim 1, wherein the instance segmentation and attribute extraction of rice grain unit located on the server and the cloud-based processing unit. The quality inspection system for rice using artificial intelligence technology, according to the principles stated in claim 1 and 2, wherein using sample rice grain images (310) with a black background.
PCT/TH2023/050008 2022-06-16 2023-06-16 Quality inspection system for rice using artificial intelligence technology WO2023244184A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
TH2203001502 2022-06-16
TH2203001502 2022-06-16

Publications (1)

Publication Number Publication Date
WO2023244184A1 true WO2023244184A1 (en) 2023-12-21

Family

ID=89191749

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/TH2023/050008 WO2023244184A1 (en) 2022-06-16 2023-06-16 Quality inspection system for rice using artificial intelligence technology

Country Status (1)

Country Link
WO (1) WO2023244184A1 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050074146A1 (en) * 2003-09-17 2005-04-07 Advanta Technology, Ltd. Method and apparatus for analyzing quality traits of grain or seed
CN101701916A (en) * 2009-12-01 2010-05-05 中国农业大学 Method for quickly identifying and distinguishing variety of corn
CN105139405A (en) * 2015-09-07 2015-12-09 中国人民解放军理工大学 Visual separating and detection method of overlapping broken grain and whole grain
CN114612692A (en) * 2022-03-04 2022-06-10 温州职业技术学院 Feature extraction method based on artificial intelligence image recognition

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050074146A1 (en) * 2003-09-17 2005-04-07 Advanta Technology, Ltd. Method and apparatus for analyzing quality traits of grain or seed
CN101701916A (en) * 2009-12-01 2010-05-05 中国农业大学 Method for quickly identifying and distinguishing variety of corn
CN105139405A (en) * 2015-09-07 2015-12-09 中国人民解放军理工大学 Visual separating and detection method of overlapping broken grain and whole grain
CN114612692A (en) * 2022-03-04 2022-06-10 温州职业技术学院 Feature extraction method based on artificial intelligence image recognition

Similar Documents

Publication Publication Date Title
Kılıç et al. A classification system for beans using computer vision system and artificial neural networks
Venora et al. Identification of Italian landraces of bean (Phaseolus vulgaris L.) using an image analysis system
Zhang et al. Detection and classification of potato defects using multispectral imaging system based on single shot method
CN105009731B (en) Corn seed investigating method and its system
Sharma et al. Grain quality detection by using image processing for public distribution
Najeeb et al. Dates maturity status and classification using image processing
Sharma et al. Image processing techniques to estimate weight and morphological parameters for selected wheat refractions
CN106645180A (en) Method for checking defects of substrate glass, field terminal and server
Ünal et al. Classification of hazelnut kernels with deep learning
Payman et al. Development of an expert vision-based system for inspecting rice quality indices
CN116559111A (en) Sorghum variety identification method based on hyperspectral imaging technology
Sidnal et al. Grading and quality testing of food grains using neural network
Gurubelli et al. Texture and colour gradient features for grade analysis of pomegranate and mango fruits using kernel-SVM classifiers
Angelia et al. Fermentation level classification of cross cut cacao beans using k-NN algorithm
CN111680738A (en) Screening device for apple quality detection and detection method thereof
Tu et al. AIseed: An automated image analysis software for high-throughput phenotyping and quality non-destructive testing of individual plant seeds
Janardhana et al. Computer aided inspection system for food products using machine vision—a review
WO2023244184A1 (en) Quality inspection system for rice using artificial intelligence technology
Agustin et al. Automatic milled rice quality analysis
Pabamalie et al. An intelligent rice quality classifier
Ayitenfsu Method of coffee bean defect detection
Chen et al. Detect black germ in wheat using machine vision
Kotwaliwale et al. Machine vision for characterisation of some phenomic features of plant parts in distinguishing varieties-a review
Kaur et al. Automatic classification of turmeric rhizomes using the external morphological characteristics
Vithu et al. Identification of dockage in paddy using multiclass SVM

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23824346

Country of ref document: EP

Kind code of ref document: A1